CN116401309B - Student personalized learning recommendation method and device based on local influence and deep preference propagation - Google Patents
Student personalized learning recommendation method and device based on local influence and deep preference propagation Download PDFInfo
- Publication number
- CN116401309B CN116401309B CN202310412943.4A CN202310412943A CN116401309B CN 116401309 B CN116401309 B CN 116401309B CN 202310412943 A CN202310412943 A CN 202310412943A CN 116401309 B CN116401309 B CN 116401309B
- Authority
- CN
- China
- Prior art keywords
- propagation
- data
- preference
- deep
- knowledge graph
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 38
- 230000005540 biological transmission Effects 0.000 claims abstract description 10
- 238000007418 data mining Methods 0.000 claims abstract description 7
- 239000013598 vector Substances 0.000 claims description 34
- 238000004364 calculation method Methods 0.000 claims description 27
- 239000011159 matrix material Substances 0.000 claims description 18
- 239000008186 active pharmaceutical agent Substances 0.000 claims description 13
- 238000012800 visualization Methods 0.000 claims description 11
- 230000006399 behavior Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 claims description 6
- 230000004044 response Effects 0.000 claims description 6
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000005516 engineering process Methods 0.000 claims description 5
- 238000004140 cleaning Methods 0.000 claims description 3
- 238000013079 data visualisation Methods 0.000 claims description 3
- 238000009792 diffusion process Methods 0.000 claims description 3
- 230000003993 interaction Effects 0.000 claims description 3
- 238000007477 logistic regression Methods 0.000 claims description 3
- 238000007781 pre-processing Methods 0.000 claims description 3
- 230000003542 behavioural effect Effects 0.000 claims description 2
- 230000000644 propagated effect Effects 0.000 claims 1
- 230000001149 cognitive effect Effects 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000001902 propagating effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Abstract
The application discloses a student personalized learning recommendation method and device based on local influence and deep preference transmission. First, data set information and knowledge-graph information are acquired using a data mining technique. And then, designing a recommendation system based on the local influence and the deep preference propagation thought, and recommending proper course learning paths and contents according to the obtained student history learning data. The recommendation system R firstly carries out preference propagation on the constructed knowledge graph, then gives local influence weights to the nodes by utilizing the node influence among the knowledge graph nodes, carries out deep preference propagation on the knowledge graph according to the node weights so as to obtain interest preference of students at a deeper level, and predicts the final click probability by utilizing the obtained node weightsAnd finally recommending proper course learning paths and contents according to the predicted click probability. Compared with the prior art, the method can effectively improve the personalized learning efficiency of students and has stronger practicability.
Description
Technical Field
The application belongs to the field of knowledge maps and recommendation systems, and particularly relates to a student personalized learning recommendation method and device based on local influence and deep preference propagation.
Background
Knowledge Graph (knowledgegraph) is a structured semantic Knowledge base used to symbolically describe concepts and their interrelationships in the physical world. The basic composition unit is an entity-relation-entity triplet, and the entities and related attribute-value pairs thereof are mutually connected through the relation to form a net-shaped knowledge structure.
The recommendation system (Recommendation System) is an algorithm system for recommending personalized content for users by analyzing information such as historical behaviors and interests of the users by utilizing technologies such as big data and machine learning. The method can help users find products, services or contents which the users may be interested in, improve user experience, and also help enterprises to improve user retention and conversion rate.
When facing to a student personalized learning recommendation method, the existing papers are mainly based on collaborative filtering recommendation algorithms based on cognitive diagnosis and resource recommendation, and the student personalized learning recommendation method mainly depends on simple cognitive diagnosis and simple recommendation of resources for students, however, the method ignores the mastering condition and learning depth of the students on learning resources, ignores deep learning preference of the students, and cannot accurately provide personalized recommendation according to the current learning condition of the students.
Disclosure of Invention
The application aims to: aiming at the problem that students neglect the mastering condition of learning resources and the learning depth in the prior art, the application provides a student personalized learning recommendation method and device based on local influence and deep preference transmission.
The technical scheme is as follows: the application provides a student personalized learning recommendation method based on local influence and deep preference transmission, which comprises the following steps:
step 1: acquiring student historical learning data D by using data mining knowledge, defining a data knowledge graph frame, and constructing a required knowledge graph G;
step 2: designing a recommendation system R based on local influence and deep preference propagation ideas, and recommending proper course learning paths and contents by using the recommendation system R according to the acquired student history learning data D; the recommendation system R firstly carries out preference propagation on the constructed knowledge graph, then gives local influence weights to the nodes by utilizing the node influence among the knowledge graph nodes, carries out deep preference propagation on the knowledge graph according to the node weights so as to obtain interest preferences of students at a deeper level, and predicts the final click probability by utilizing the obtained node weightsFinally recommending proper course learning path and inner according to predicted click probabilityA container;
step 3: the knowledge graph G is visualized to help students better understand the relationships and structure between the multi-course knowledge points.
Further, the specific method of the step 1 is as follows:
step 1.1: acquiring student history learning data D;
step 1.2: carrying out data cleaning and data preprocessing on the student history learning data D by using data mining knowledge;
step 1.3: the preprocessed data construct a required knowledge graph g= (h, r, t) through a graph database Neo4 j.
Further, in the step 2, the specific method for carrying out preference propagation on the constructed knowledge graph is as follows:
step 2.1: acquiring student set information U= { U 1 ,u 2 ,...,u n Information v= { V 1 ,v 2 ,...v m }；
Step 2.2: building a student history behavior interaction matrix Y= { Y according to student history behavior data D uv U e U, V e V, where y uv Take the value of
Step 2.3: given a student set U, a course learning set V and a knowledge graph G, a student history learning set H is formed u ＝{v|y uv Using =1 } as seed subset to make preference propagation, and making layer-by-layer diffusion from inside to outside along the entity in the knowledge graph G to obtain k (k=1, 2, …, N) jump propagation preference set of student uThe definition is as follows: />
Further, in the step 2, the node influence between the nodes of the knowledge graph is utilized to give the node local influence weight to the specific operation as follows:
step 2.4: definition of the definitionLearning preference propagation set for students
Step 2.5: knowledge graph node local influence weighting C for student preference propagation set ki The calculation is performed such that,wherein X is i (i ε k) is the node importance value.
Further, in the step 2, knowledge-graph deep preference propagation is performed according to the node weights to obtain deeper student interest preferences, and the obtained node weights are used for predicting the final click probabilityThe specific operation is as follows:
step 2.6: giving student information U, a knowledge graph G, a propagation hop count N and a propagation index DS;
step 2.7: define N to loop from 0 to N, if N is 0 user U i As the head vector Su of the next propagation n [head]Otherwise, obtaining the last propagation tail vector Su n-1 [tail]DS vectors with the largest local influence in the vector are taken as head vectors Su of the next propagation n [head]Traversing the knowledge graph G through Su n [head]Acquisition of Su n [relation]With Su n [tail]Su is calculated by utilizing local influence algorithm n Middle node influence value weight C ki And is connected with Su n [head]Multiplying and assigning Su n [head]At this time, whether the number of the current propagation vectors is smaller than DS is judged, if so, data completion operation is carried out, namely, the data are randomly and repeatedly assigned, otherwise, the last propagation tail vector Su is obtained n-1 [tail]DS vectors with the largest local influence in the vector are taken as head vectors Su of the next propagation n [head]Finally, the current propagation vector set Su is obtained n (Su n [head],Su n [relation],Su n [tail]) The method comprises the steps of carrying out a first treatment on the surface of the Ending the loop and finally obtaining deep propagation preference data set deep S of user i ；
Step 2.8: embedding course school information V to obtain course information matrix V e By deep S i Head node h in triplet, local influence value C of h ki Head node h after weighting is calculated as a weight matrix i The calculation formula is h i ＝C ki h；
Step 2.9: by deep S i Relation node r and node h in triplet i Embedding matrix V with user e Performing a calculation to obtain a result of the entity v and the head node h i The correlation probability p under the influence of the relation r i The specific calculation process is p i ＝softmax(v T r i h i ) Wherein v is T For the transposed matrix of the entity v, softmax is the normalized exponential function and its calculation formula is
Step 2.10: will calculate the associated probability p i And tail node t in triplet i Matrix multiplication is performed to obtain a primary response of user u with respect to entity vThe calculation formula is +.>
Step 2.11: transfer of user interest sets from historical behavioral data to deep preference propagation first propagation setWill be given by equation p i ＝softmax(v T r i h i ) The entity v in (2) is replaced by->And adding 1 to the hop count, repeating the deep preference propagation process to obtain a secondary response +.>The embedded representation u of user u with respect to entity v can be obtained through successive iterations of hop counts e The calculation formula is ∈>
Step 2.12: embedding u by user e Outputting predicted click probabilities in conjunction with item embedding vThe calculation formula is as follows: />Wherein sigmoid is a logistic regression function with a calculation formula of +.>
Further, the specific method of the step 3 is as follows:
step 3.1: constructing a front-end knowledge graph visualization interface F by using Apache ECharts and a data visualization chart library and combining front-end HTML, CSS and javascript technologies;
step 3.2: establishing a background data propagation channel by using a Python Web application framework Django, constructing a data transmission port by combining a REST framework, and transmitting knowledge graph data in a database to the port;
step 3.3: and the front-end knowledge graph visualization interface F receives the port information and displays the data to the knowledge graph visualization interface F.
The application also discloses a student personalized learning recommendation device based on the local influence and the deep preference propagation, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program is loaded to the processor to execute the steps of the student personalized learning recommendation method based on the local influence and the deep preference propagation.
The beneficial effects are that:
1. the method is based on the existing neo4j graph relational database as data storage, and obtains the deep interest expression of the user by utilizing the knowledge graph deep preference propagation mode, so that the interest characteristics of the user are extracted more comprehensively, and more accurate user portraits are obtained. The knowledge graph deep preference transmission process adopts the local influence weight as the node weight, so that the problem of the utilization rate of the knowledge graph nodes is effectively solved, and meanwhile, the local influence weight value of the nodes is used as the deep preference transmission basis, so that the recommendation efficiency is improved.
2. The application can also use the visualization technology to carry out the visualization analysis on the knowledge graph, thereby helping students to better understand the relationship and structure among the knowledge points of multiple courses.
Drawings
FIG. 1 is a general flow chart of the present application;
FIG. 2 is a schematic diagram of knowledge-graph preference propagation;
FIG. 3 is a schematic of deep preference propagation with local influence weights added;
FIG. 4 is a schematic diagram of a general recommendation method according to the present application;
FIG. 5 is a flow chart of an embodiment of the present application.
Detailed Description
The present application is further illustrated below in conjunction with specific embodiments, it being understood that these embodiments are meant to be illustrative of the application and not limiting the scope of the application, and that modifications of the application, which are equivalent to those skilled in the art to which the application pertains, fall within the scope of the application defined in the appended claims after reading the application.
The application discloses a student personalized learning recommendation method and device based on local influence and deep preference transmission, wherein the student personalized learning recommendation method based on local influence and deep preference transmission comprises the following steps:
step 1: and acquiring student historical learning data D by using data mining knowledge, defining a data knowledge graph frame, and constructing a required knowledge graph G.
Step 1.1: and acquiring student historical learning data D through a school educational administration system and an online course website.
Step 1.2: and carrying out data cleaning and data preprocessing on the student historical learning data D by using the data mining knowledge.
Step 1.3: the preprocessed data construct a required knowledge graph g= (h, r, t) through a graph database Neo4 j.
Step 2: and designing a recommendation system R based on the local influence and deep preference propagation ideas, and recommending proper course learning paths and contents according to the obtained student history learning data D.
Step 2.1: acquiring student set information U= { U 1 ,u 2 ,...,u n }。
Step 2.2: acquiring curriculum academic information v= { V 1 ,v 2 ,...v m }。
Step 2.3: building a student history behavior interaction matrix Y= { Y according to student history behavior data D uv U e U, V e V, where y uv Take the value of
Step 2.4: given a student set U, a course learning set V and a knowledge graph G, a student history learning set H is formed u ＝{v|y uv Using =1 } as seed subset to make preference propagation, and making layer-by-layer diffusion from inside to outside along the entity in the knowledge graph G to obtain k (k=1, 2, …, N) jump propagation preference set of student uThe definition is as follows: />A knowledge-graph preference propagation schematic is shown in fig. 2. As the number of propagation times increases, more and more nodes join the propagation preference set. In the figure, the white blank portion represents nodes that have joined the propagation preference set. The gray part indicates the node currently propagating, i.e. the node that needs neighbor propagation.
Step 2.5: defining a learning preference propagation set for students
Step 2.6: knowledge graph node local influence weighting C for student preference propagation set ki The calculation is performed such that,wherein X is i (i ε k) is the node importance value.
Step 2.7: given student information U, knowledge graph G, propagation hop count N, propagation index DS.
Step 2.8: define N to loop from 0 to N, if N is 0, user U i As the head vector Su of the next propagation n [head]Otherwise, obtaining the last propagation tail vector Su n-1 [tail]DS vectors with the largest local influence in the vector are taken as head vectors Su of the next propagation n [head]Traversing the knowledge graph G through Su n [head]Acquisition of Su n [relation]With Su n [tail]Su is calculated by utilizing local influence algorithm n Middle node influence value weight C ki And is connected with Su n [head]Multiplying and assigning Su n [head]At this time, whether the number of the current propagation vectors is smaller than DS is judged, if so, data completion operation is carried out, namely, the data are randomly and repeatedly assigned, otherwise, the last propagation tail vector Su is obtained n-1 [tail]DS vectors with the largest local influence in the vector are taken as head vectors Su of the next propagation n [head]Finally, the current propagation vector set Su is obtained n (Su n [head],Su n [relation],Su n [tail]). Ending the loop and finally obtaining deep propagation preference data set deep S of user i . A schematic of deep preference propagation with local influence weights added is shown in fig. 3. In the figure, the shades of elliptical color represent the nodes involved in each propagation process. The white circles represent nodes that do not propagate deeply because their local impact weights are too low and the number of current propagation vectors exceeds the propagation index DS.
Step 2.9: embedding information V such as courses to obtain course information matrix V e By deep S i Head node h in triplet, hLocal influence value C ki Head node h after weighting is calculated as a weight matrix i The calculation formula is h i ＝C ki h。
Step 2.10: by deep S i Relation node r and node h in triplet i Embedding matrix V with user e Performing a calculation to obtain a result of the entity v and the head node h i The correlation probability p under the influence of the relation r i The specific calculation process is p i ＝softmax(v T r i h i ) Wherein v is T For the transposed matrix of the entity v, softmax is the normalized exponential function and its calculation formula is
Step 2.11: will calculate the associated probability p i And tail node t in triplet i Matrix multiplication is performed to obtain a primary response of user u with respect to entity vThe calculation formula is +.>
Step 2.12: the user interest set at this point has been transferred from the historical behavior data to the deep preference propagation first propagation setWill be given by equation p i ＝softmax(v T r i h i ) The entity v in (2) is replaced by->And adding 1 to the hop count, repeating the deep preference propagation process to obtain a secondary response +.>The embedded representation u of user u with respect to entity v can be obtained through successive iterations of hop counts e . The calculation formula is +.>
Step 2.13: embedding u by user e Outputting predicted click probabilities in conjunction with item embedding vThe calculation formula is as follows: />Wherein sigmoid is a logistic regression function with a calculation formula of +.>
Step 3: in addition, knowledge graph G can be visualized to help students better understand relationships and structures between multi-course knowledge points. In specific implementation, the method can be realized by the following steps:
step 3.1: constructing a front-end knowledge graph visualization interface F by using Apache ECharts and a data visualization chart library and combining front-end HTML, CSS and javascript technologies;
step 3.2: and establishing a background data propagation channel by using a Python Web application framework Django, constructing a data transmission port by combining a REST framework, and transmitting knowledge graph data in a database to the port.
Step 3.3: and the front-end knowledge graph visualization interface F receives the port information and displays the data to the knowledge graph visualization interface F.
The relevant letters involved in the above steps are explained as follows:
the application discloses a student personalized learning recommendation device based on local influence and deep preference propagation, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein the computer program is loaded to the processor to execute the steps of the student personalized learning recommendation method based on local influence and deep preference propagation.
The foregoing embodiments are merely illustrative of the technical concept and features of the present application, and are intended to enable those skilled in the art to understand the present application and to implement the same, not to limit the scope of the present application. All equivalent changes or modifications made according to the spirit of the present application should be included in the scope of the present application.
Claims (4)
1. A student personalized learning recommendation method based on local influence and deep preference propagation is characterized by comprising the following steps:
step 1: acquiring student historical learning data D by using data mining knowledge, defining a data knowledge graph frame, and constructing a required knowledge graph G;
step 2: designing a recommendation system R based on local influence and deep preference propagation ideas, and recommending proper course learning paths and contents by using the recommendation system R according to the acquired student history learning data D; the recommendation system R firstly carries out preference propagation on the constructed knowledge graph, then gives local influence weights to the nodes by utilizing the node influence among the knowledge graph nodes, carries out deep preference propagation on the knowledge graph according to the node weights so as to obtain interest preferences of students at a deeper level, and predicts the final click probability by utilizing the obtained node weightsFinally recommending proper course learning paths and contents according to the predicted click probability;
the specific method for carrying out preference propagation on the constructed knowledge graph comprises the following steps:
step 2.1: acquiring student set information U= { U 1 ,u 2 ,...,u n ' lessons }, lessonsProgramming information V= { V 1 ,v 2 ,...v m -a }; step 2.2: building a student history behavior interaction matrix Y= { Y according to student history behavior data D uv U e U, V e V, where y uv Take the value ofStep 2.3: given a student set U, a course learning set V and a knowledge graph G, a student history learning set H is formed u ＝{v|y uv Using =1 } as seed set to make preference propagation, and making layer-by-layer diffusion along the entity in the knowledge graph G from inside to outside so as to obtain k-hop propagation preference set +.>The definition is as follows:wherein k=1, 2, …, N;
the node influence among the knowledge graph nodes is utilized to endow the nodes with local influence weight specific operations as follows: step 2.4: defining a learning preference propagation set for studentsStep 2.5: knowledge graph node local influence weighting C for student preference propagation set ki Calculation of->Wherein X is i (i ε k) is the node importance value;
carrying out deep preference propagation of knowledge patterns according to node weights to obtain interest preferences of students at deeper levels, and predicting final click probability by using the obtained node weightsThe specific operation is as follows:
step 2.6: giving student information U, a knowledge graph G, a propagation hop count N and a propagation index DS; step 2.7: define N to loop from 0 to N, if N is 0 user U i Historical behavior data of (a)For the next propagated head vector Su n [head]Otherwise, obtaining the last propagation tail vector Su n-1 [tail]DS vectors with the largest local influence in the vector are taken as head vectors Su of the next propagation n [head]Traversing the knowledge graph G through Su n [head]Acquisition of Su n [relation]With Su n [tail]Su is calculated by utilizing local influence algorithm n Middle node influence value weight C ki And is connected with Su n [head]Multiplying and assigning Su n [head]At this time, whether the number of the current propagation vectors is smaller than DS is judged, if so, data completion operation is carried out, namely, the data are randomly and repeatedly assigned, otherwise, the last propagation tail vector Su is obtained n-1 [tail]DS vectors with the largest local influence in the vector are taken as head vectors Su of the next propagation n [head]Finally, the current propagation vector set Su is obtained n (Su n [head],Su n [relation],Su n [tail]) The method comprises the steps of carrying out a first treatment on the surface of the Ending the loop and finally obtaining deep propagation preference data set deep S of user i ；
Step 2.8: embedding course school information V to obtain course information matrix V e By deep S i Head node h in triplet, local influence value C of h ki Head node h after weighting is calculated as a weight matrix i The calculation formula is h i ＝C ki h；
Step 2.9: by deep S i Relation node r and node h in triplet i Embedding matrix V with user e Performing a calculation to obtain a result of the entity v and the head node h i The correlation probability p under the influence of the relation r i The specific calculation process is p i ＝softmax(v T r i h i ) Wherein v is T For the transposed matrix of the entity v, softmax is the normalized exponential function and its calculation formula is
Step 2.10: will calculate the associated probability p i And tail node t in triplet i Matrix multiplication is performed to obtain a primary response of user u with respect to entity vThe calculation formula is +.>
Step 2.11: transfer of user interest sets from historical behavioral data to deep preference propagation first propagation setWill be given by equation p i ＝softmax(v T r i h i ) The entity v in (2) is replaced by->And adding 1 to the hop count, repeating the deep preference propagation process to obtain a secondary response +.>The embedded representation u of user u with respect to entity v can be obtained through successive iterations of hop counts e The calculation formula is ∈>
Step 2.12: embedding u by user e Outputting predicted click probabilities in conjunction with item embedding vThe calculation formula is as follows:wherein sigmoid is a logistic regression function with a calculation formula of +.>
Step 3: the knowledge graph G is visualized to help students better understand the relationships and structure between the multi-course knowledge points.
2. The personalized learning recommendation method for students based on local influence and deep preference propagation according to claim 1, wherein the specific method of step 1 is as follows:
step 1.1: acquiring student history learning data D;
step 1.2: carrying out data cleaning and data preprocessing on the student history learning data D by using data mining knowledge;
step 1.3: the preprocessed data construct a required knowledge graph g= (h, r, t) through a graph database Neo4 j.
3. The personalized learning recommendation method for students based on local influence and deep preference propagation according to claim 1, wherein the specific method in the step 3 is as follows:
step 3.1: constructing a front-end knowledge graph visualization interface F by using Apache ECharts and a data visualization chart library and combining front-end HTML, CSS and javascript technologies;
step 3.2: establishing a background data propagation channel by using a Python Web application framework Django, constructing a data transmission port by combining a REST framework, and transmitting knowledge graph data in a database to the port;
step 3.3: and the front-end knowledge graph visualization interface F receives the port information and displays the data to the knowledge graph visualization interface F.
4. A student personalized learning recommendation device based on local influence and deep preference propagation, comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the computer program when loaded to the processor performs the steps of a student personalized learning recommendation method based on local influence and deep preference propagation as claimed in any one of claims 1-3.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310412943.4A CN116401309B (en) | 2023-04-18 | 2023-04-18 | Student personalized learning recommendation method and device based on local influence and deep preference propagation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310412943.4A CN116401309B (en) | 2023-04-18 | 2023-04-18 | Student personalized learning recommendation method and device based on local influence and deep preference propagation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116401309A CN116401309A (en) | 2023-07-07 |
| CN116401309B true CN116401309B (en) | 2023-09-26 |
Family
ID=87015766
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310412943.4A Active CN116401309B (en) | 2023-04-18 | 2023-04-18 | Student personalized learning recommendation method and device based on local influence and deep preference propagation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116401309B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103379158A (en) * | 2012-04-24 | 2013-10-30 | 中国移动通信集团公司 | Method and system for recommending friend information in social network |
| CN108280114A (en) * | 2017-07-28 | 2018-07-13 | 淮阴工学院 | A kind of user's literature reading interest analysis method based on deep learning |
| CN113158033A (en) * | 2021-03-19 | 2021-07-23 | 浙江工业大学 | Collaborative recommendation model construction method based on knowledge graph preference propagation |
| CN114510653A (en) * | 2022-04-21 | 2022-05-17 | 中国科学技术大学 | Social group recommendation method, system, device and storage medium |
| CN115146161A (en) * | 2022-06-30 | 2022-10-04 | 武汉美和易思数字科技有限公司 | Personalized learning resource recommendation method and system based on content recommendation |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140272889A1 (en) * | 2013-03-15 | 2014-09-18 | Career Education Center | Computer implemented learning system and methods of use thereof |
-
2023
- 2023-04-18 CN CN202310412943.4A patent/CN116401309B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103379158A (en) * | 2012-04-24 | 2013-10-30 | 中国移动通信集团公司 | Method and system for recommending friend information in social network |
| CN108280114A (en) * | 2017-07-28 | 2018-07-13 | 淮阴工学院 | A kind of user's literature reading interest analysis method based on deep learning |
| CN113158033A (en) * | 2021-03-19 | 2021-07-23 | 浙江工业大学 | Collaborative recommendation model construction method based on knowledge graph preference propagation |
| CN114510653A (en) * | 2022-04-21 | 2022-05-17 | 中国科学技术大学 | Social group recommendation method, system, device and storage medium |
| CN115146161A (en) * | 2022-06-30 | 2022-10-04 | 武汉美和易思数字科技有限公司 | Personalized learning resource recommendation method and system based on content recommendation |
Non-Patent Citations (5)
| Title |
|---|
| An Intelligent Context-Aware Management Framework for Cold Chain Logistics Distribution;Xiang Li等;《Web of Science》;全文 * |
| Recommendation in Event-Based Social Networks.《Springer》.2015,全文. * |
| Semantic-Enhanced and Context-Aware Hybrid Collaborative Filtering for Event Recommendation in Event-Based Social Networks;MINGHUA XU等;《IEEE》;全文 * |
| 基于用户活跃度和影响力的社区推荐系统关键技术的研究;于笑明;《信息科技》;全文 * |
| 基于社会影响力的网络用户推荐方法研究;王欢;《信息科技》;全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116401309A (en) | 2023-07-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jain et al. | Discover opinion leader in online social network using firefly algorithm | |
| Kuipers et al. | Partition MCMC for inference on acyclic digraphs | |
| Rahwan et al. | Coalition structure generation: A survey | |
| CN108804689B (en) | Question-answering platform-oriented label recommendation method integrating user hidden connection relation | |
| Hou et al. | Course recommendation of MOOC with big data support: A contextual online learning approach | |
| Bellot et al. | NetBenchmark: a bioconductor package for reproducible benchmarks of gene regulatory network inference | |
| Hahn et al. | Truth tracking performance of social networks: How connectivity and clustering can make groups less competent | |
| Cho et al. | Reconstructing causal biological networks through active learning | |
| Gan et al. | Knowledge structure enhanced graph representation learning model for attentive knowledge tracing | |
| CN103577876B (en) | Based on feedforward neural network credible with insincere user identification method | |
| CN107391542A (en) | A kind of open source software community expert recommendation method based on document knowledge collection of illustrative plates | |
| Lin et al. | Steering information diffusion dynamically against user attention limitation | |
| Yang et al. | Recommender system-based diffusion inferring for open social networks | |
| Chen et al. | A temporal recommendation mechanism based on signed network of user interest changes | |
| Zhou et al. | Rank2vec: learning node embeddings with local structure and global ranking | |
| Wangkhamhan | Adaptive chaotic satin bowerbird optimisation algorithm for numerical function optimisation | |
| Ke et al. | Sparse attentive backtracking: Long-range credit assignment in recurrent networks | |
| Lin | Learning information recommendation based on text vector model and support vector machine | |
| CN112052995B (en) | Social network user influence prediction method based on fusion emotion tendency theme | |
| CN116401309B (en) | Student personalized learning recommendation method and device based on local influence and deep preference propagation | |
| CN115730752A (en) | Self-adaptive learning path planning method based on knowledge interest network | |
| CN115827968A (en) | Individualized knowledge tracking method based on knowledge graph recommendation | |
| CN106372147B (en) | Heterogeneous topic network construction and visualization method based on text network | |
| Nithya et al. | Optimization-assisted personalized event recommendation for event-based social networks | |
| Lin et al. | A genetic algorithm approach for detecting hierarchical and overlapping community structure in dynamic social networks |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |